Predicting the Usefulness of Collection Enrichment for Enterprise Search

Authors:
Jie Peng;Ben He;Iadh Ounis
Affiliations:
Department of Computing Science, University of Glasgow, UK G12 8QQ;Department of Computing Science, University of Glasgow, UK G12 8QQ;Department of Computing Science, University of Glasgow, UK G12 8QQ
Venue:
ICTIR '09 Proceedings of the 2nd International Conference on Theory of Information Retrieval: Advances in Information Retrieval Theory
Year:
2009

Citing 5
Cited 2

Improving two-stage ad-hoc retrieval for short queries

Proceedings of the 21st annual international ACM SIGIR conference on Research and development in information retrieval
Predicting query performance

SIGIR '02 Proceedings of the 25th annual international ACM SIGIR conference on Research and development in information retrieval
Improving the estimation of relevance models using large external corpora

SIGIR '06 Proceedings of the 29th annual international ACM SIGIR conference on Research and development in information retrieval
Query performance prediction

Information Systems
University of glasgow at WebCLEF 2005: experiments in per-field normalisation and language specific stemming

CLEF'05 Proceedings of the 6th international conference on Cross-Language Evalution Forum: accessing Multilingual Information Repositories

Learning models for ranking aggregates

ECIR'11 Proceedings of the 33rd European conference on Advances in information retrieval
Learning to select a ranking function

ECIR'2010 Proceedings of the 32nd European conference on Advances in Information Retrieval

Quantified Score

Hi-index	0.00

Visualization

Abstract

Query Expansion (QE) often improves the retrieval performance of an Information Retrieval (IR) system. However, as enterprise intranets are often sparse in nature, with limited use of alternative lexical representations between authors, it can be advantageous to use Collection Enrichment (CE) to gather higher quality pseudo-feedback documents. In this paper, we propose the use of query performance predictors to selectively apply CE on a per-query basis. We thoroughly evaluate our approach on the CERC standard test collection and its corresponding topic sets from the TREC 2007 & 2008 Enterprise track document search tasks. We experiment with 3 different external resources and 3 different query performance predictors. Our experimental results demonstrate that our proposed approach leads to a significant improvement in retrieval performance.